|
CDR Burn Speed and Audio Quality
|
|
|
rgoodwin
Newbie
|
18. December 2002 @ 00:30 |
Link to this message
|
|
I've always assumed that reducing the burn speed of a CDR increased the quality - that also seems to be the advice offered on forums such as this.
When I've mastered Audio CDs to be sent to pressing plants, I've always written them single-speed etc etc.
I recently purchased some TDK d-view CD-Rs (details at the bottom of this post), and recorded them single-speed using CDR-Win and a Yamaha CRW6416S scsi drive.
The audio quality when played back on both domestic and car CD players was terrible - lots of crackle and distortion (worse on the car player than the home hi-fi).
So, as a test, I recorded one at a faster speed (6x).
Result: Sounds perfect both on the Hi Fi and in the Car.
Has anybody else observed this behaviour, or can comment on why this might be?
Regards
Richard Goodwin
CDR Details:
ATIP: 97m 24s 01f
Disc Manufacturer: Taiyo Yuden Company Ltd.
Reflective layer: Dye (Long strategy; e.g. Cyanine, Azo etc.)
Media type: CD-Recordable
Recording Speeds: min. unknown - max. unknown
nominal Capacity: 702.83MB (79m 59s 73f / LBA: 359848)
|
|
Advertisement
|
 |
|
|
|
cd-rw.org
Senior Member
4 product reviews
|
18. December 2002 @ 11:33 |
Link to this message
|
|
To tell you the truth, I have a hard time beleiving that write speed would be the cause. Are you able to re-produce the bad sound to another disc? Are you sure that the audio source is ok?
BUT, it has been reported (by Ritek actually) that very low write speed may cause pit smearing, and therefore reduce quality. But that the difference would be of this magnitude...hm..
Does anyone here have Tayos? Can anyone else get similar results.
|
|
rgoodwin
Newbie
|
18. December 2002 @ 21:15 |
Link to this message
|
|
The Audio is fine - I run a small recording studio, so I know that side of it is OK. All these tests were made with the same Audio files.
I burnt about 20 of the TDK's at single speed before I noticed, I didn't play every single one, I tried several, and they were all the same.
I've also got a spindle-pack of Durabrand CDRs (Details below), and they seem to exhibit the same behaviour.
I stopped using the Durabrands a while back because I'd noticed that they distorted the audio, and wouldn't play reliably in either my car, or on a CD Walkman, though they were OK on the home Hi Fi.
I've just tried recording one of them at 6-speed, and it sounds fine, so same behaviour.
Is there any software around that'll indicate the Bit Error Rate of a recorded CD?
A few years back I had a pressing plant refuse to press a disk because they said the Error Rate was too high, so they must have measured it somehow.
Ever since then I've always written pressing plant masters at single speed and used known good quality blanks, and never had any problems (that I know of!).
I've had the CDR/W Drive a while now (a couple of years, I think).
Is it possible that the drive has degraded in some way? Could the laser have got dust on it or something?
Is there a good way of cleaning it without dismantling the drive?
Richard.
Details of Durabrand CD-R
ATIP: 97m 26s 66f
Disc Manufacturer: CMC Magnetics Corp.
Reflective layer: Dye (Short strategy; e.g. Phthalocyanine)
Media type: CD-Recordable
Recording Speeds: min. unknown - max. unknown
nominal Capacity: 702.83MB (79m 59s 74f / LBA: 359849)
|
|
Pio2001
Moderator
|
20. December 2002 @ 08:28 |
Link to this message
|
Yes, this behaviour have already been reported. It seems less common than the opposite. But the mainstream idea is that there is no fixed behaviour, it depends on the burner and the media.
Examples :
http://www.emedialive.com/EM2000/starrett5.html shows the same behaviour from a Memorex CDR and Plextor burners, and the opposite with Imation and Mitsui CDRs on the same burner. Sony and TDK behaving chaoticly, with no clear improvement or decaying with speed.
Note that in theory, the numbers show a perfect quality at any speed and any media, because an error rate of 0.3 instead of 1.5 when 50 is considered good, and 150 bad, is negligible.
However, it may make the difference on CD Players not officially supporting CDR playback.
Another example : http://www-user.tu-chemnitz.de/~noe/LiteOn/C1-scans/index-eng.html , reports the LiteOn LTR-40125S/W @ 48125W, VS02 burner to be unable to burn slower than 24x properly, exept with Verbatim CDRs.
Pio2001
|
|
cd-rw.org
Senior Member
4 product reviews
|
20. December 2002 @ 12:51 |
Link to this message
|
|
Ok, so this is an extreme case of pit smearing.
Pio,
You talk about Plextor, LiteOn and TDK. Early TDKs were Plextors, then they changed to Sanyo (that also makes parts for Plextor) and finally TDK has been rebadging LiteOns.
rgoodwin,
Perhaps Yamaha F1 could be something for you? It's audio master technilogy aims at minimizing the error rate by using 1) low write speeds 2) increasing the length of pits and gaps (by using a slightly higher rotation speed)
|
|
Pio2001
Moderator
|
21. December 2002 @ 10:53 |
Link to this message
|
|
No, no, TDK CDRs, not drives
|
|
cd-rw.org
Senior Member
4 product reviews
|
21. December 2002 @ 12:20 |
Link to this message
|
|
Oh, sorry Pio - my bad.
Well, they still reference the disk simply as "TDK". It could be: TDK Corp, T.Y., Ritek, CMC...
|
|
rgoodwin
Newbie
|
22. December 2002 @ 05:09 |
Link to this message
|
|
>>Perhaps Yamaha F1 could be something for you? It's audio master technilogy aims at minimizing the error rate by using 1) low write speeds 2) increasing the length of pits and gaps (by using a slightly higher rotation speed)
That sounds interesting. If I continue having problems, I'll certainly consider replacing the drive.
When you've put many hundreds of hours of your life into a CD, you can't risk sending the pressing plant a dud master.
Richard
|
|
Wildstar
Junior Member
|
22. December 2002 @ 05:52 |
Link to this message
|
|
I have a Yamaha 4416E, and it doesn't like 80 min CD-Rs... The write process finishes ok, but the quality leaves a lot to be desired, especially when using 1x speed. In fact, 2x speed seems to give me better sound quality and data readability (not 4x, though). It is my opinion that these old Yamaha drives were optimised for 2x recording, and thus function best at this speed. You should try burning the same CD at different speeds and comparing the results. Also note that these Yamahas have been known to start making some tracking errors, as they "age". My drive could never DAE properly anyway (this shows you how good it is in audio work...)
I got a Plextor 4012A to replace it, and its audio recordings sound much better that the Yamaha's, so if you use yours for audio mastering, i recommend you get a new burner ;)
|
|
cd-rw.org
Senior Member
4 product reviews
|
22. December 2002 @ 08:25 |
Link to this message
|
|
Wildstar,
Also, old Yamaha burners had a reputation for being extremely picky about media. I have no facts to back this up with, but that's what I remember being discussed years ago. My friend has a 4x Yamaha SCSI writer and it certainly was a picky device.
|
|
aqua
Suspended due to non-functional email address
|
22. December 2002 @ 20:47 |
Link to this message
|
|
cd-rw.org,
you tell about the audio master technology of yamaha increase the length of pits and gaps. Does it mean that you will a 80 min CD-R won't be able to really have 80 min of music becasue de pits and gaps are longer ? I have seen a yamaha cd writer box today and I understand how it works but don't understand if you can still put 80 min of music on a 80 min CD-R.
|
|
rgoodwin
Newbie
|
22. December 2002 @ 21:27 |
Link to this message
|
|
Pio2001 Said:
>>Note that in theory, the numbers show a perfect quality at any speed and any media, because an error rate of 0.3 instead of 1.5 when 50 is considered good, and 150 bad, is negligible.
Is there a way that I can measure the error rate that I'm getting from various different burn speeds?
Richard
|
|
Wildstar
Junior Member
|
23. December 2002 @ 04:04 |
Link to this message
|
|
|
|
Pio2001
Moderator
|
31. December 2002 @ 15:13 |
Link to this message
|
Pio2001
This message has been edited since posting. Last time this message was edited on 31. December 2002 @ 15:14
|
|
rgoodwin
Newbie
|
2. January 2003 @ 05:11 |
Link to this message
|
|
Many thanks for the pointer to the writing quality article.
I've downloaded it, and will read it with interest.
Richard
|
|
mr_lemon
Junior Member
|
19. January 2003 @ 05:04 |
Link to this message
|
|
Hi,
I also do Audio Mastering, with a Yamaha 8424S SCSI burner, and now an LG 8400B 40Xburner.
With the Yamaha, I found that any burn speed above 1x would really kill the audio quality. I am not talking about distortion, just the way the sound changed(for the negative) at the higher burn speeds. I always burned at 1x with the Yamaha, with near perfect results. Mind u, it was a newer drive then the one u listed, so might have incorporated some fixes for CD-R compatibility.
My LG 8400B does not change the sound quality much at all. Seems to make 'almost' perfect copies from the original at just about any speed. This burner's lowest speed is 8x and I get great masters. I can still hear a slight difference between the original and burn, but its so small, I have a hard time figuring out if its a bad thing.
I think newer drives are just better at getting the pits and lands correct. Even at high speeds. I can only speak about the LG burner, but i am thinking any drive designed to burn at 40x will have to have a much more accurate laser tracking system, thus giving u better burns at most speeds.
TOM...
|
|
cd-rw.org
Senior Member
4 product reviews
|
19. January 2003 @ 05:23 |
Link to this message
|
|
Mr lemon,
You do realise that if the sound quality would differ, that would mean that bits are changed in the process? Or alternatively, the used player is unable to read from the CD-R's and applies error correction schemes.
Have you made bit-by-bit comparisons of burned sound files? (note that CD-R drive offsets cause differencesin results), or performed reliable blind listening tests?
|
|
mr_lemon
Junior Member
|
19. January 2003 @ 05:47 |
Link to this message
|
|
Hi,
I have not done a bit for bit comparison of the data on the disc. Maybe the data is bit-for-bit identical, but the burned disc might have problems reading on Audio CD players. Whatever the reason, the different burn speeds on my Yamaha 8424S where clearly audibly.
Thinks like missing bass end and exagerated treble content can't be missed between 1x and 8x on my Yamaha. Not much difference on the LG 8400B even at 24x.
I have a feeling the Yamaha just made it difficult to SEE/READ the audio CD's on regular cd players. The CD player most likely was error correcting overtime which leads to the sound change.
As for blind listening tests, not official results, but the difference was so obvious. I have been an Audiophile for 14years, and have spent most of that time also doing Recording and Mixing in a home studio. Maybe i just hear more problems then most people would, or maybe my stero system is much more resolving of the differences.
Anyways, the difference with my Yamaha, was unexceptable to me, but might be resonable results for most people, but i doubt if someone cared about sound quality, that they would not hear the difference between 1x and 8x on my Yamaha 8424S.
Oh and it did sound better with Black discs too, but again that might be an error correcting issue on playback, related to laser light scatter/noise perhaps??
TOM....
|
|
cd-rw.org
Senior Member
4 product reviews
|
19. January 2003 @ 13:23 |
Link to this message
|
|
Black disc are normal phthalocyanine discs. The wavelenght of the laser is different form visible light. Therefore the 'blackness' of the disc is apparent for our eyes only, and trasparent to a laser beam.
With all the respect, I have heard this kind of stament more times that I can remember. They very seldom are proved in any way. Therefore I am sceptic towards these results of yours, unless they are backed by undeniable results or a good theory.
The theory of CD player constantly error correcting is one possibility. On the other hand, it sound illogical what you say about bass and treble, regarding that it would be due to error correction.
Old school audiophile here too, been for 10 years or so. Running Genelec studio monitors here at home.
This message has been edited since posting. Last time this message was edited on 19. January 2003 @ 13:27
|
|
mr_lemon
Junior Member
|
19. January 2003 @ 13:45 |
Link to this message
|
|
Yello,
I had heard somewhere that the laser gets scattered and reflected back into the pick up lense a lot more in clear disks, then black discs. I have not really read any proof on this speculation, but it sounded good at the time.
Well all the proof i need for my results are my ears. The only way you could hear these differences would be for me to send u some discs.
I am not very interested in getting into audiophile debates about subjective or objective listening. I am more then convinced blind testing does not tell anything about sound. I am also convinced things like wire,capacitors, and power supplies DO make a difference in sound quality for listening or recording.
I don't know why my burner sounded worse at higer speeds, and i dont know why at the same burn speed (1x), black discs sounded better then clear discs, from the same CD-R media manufacturer even.
What I do know is, these (high speed) burned discs did not sound like the original recording or the 1x burned disc.
Jitter? Media compatibility? Bad Burns? Bad Media? What explains it?
TOM...
Im using Mission 731's for monitors in my studio and computer, and B&W CDM1's for the stereo system.
|
|
rgoodwin
Newbie
|
20. January 2003 @ 04:22 |
Link to this message
|
|
This is absolutely the same behaviour I have been observing for some time now.
Disks burned on my yamaha drive at higher speeds have a very 'fizzy' sound particularly noticeable during louder passages - like the audio file has been run through an aural exciter with all nthe knobs set to 11.
I know that the WAV file is OK, I can play it back through the optical interface on my PC straight into a digital mixer and then out to a variety of speakers. No problem.
There is definitely a clearly audible difference when CDs are burned using different write speeds.
The difference shows up more on my Car CD player and Pioneer personal Hi Fi than it does on the Denon CD player connected to my domestic Hi Fi.
My presumption is that the difference in audio quality is due to the player(s) heavily error-correcting during playback, so the different error rates MUST be being caused by the different burn speeds.
I wish there was a simple way of measuring the error rates on different disks.
Richard
|
|
cd-rw.org
Senior Member
4 product reviews
|
20. January 2003 @ 04:42 |
Link to this message
|
|
Mr lemon,
The laser scattering is a marketing mantra which was used when the black CD-R was introduced. But the black material can't absorb laser wavelenghts, since the disc wouldn't work at all if that was the case (or it would significantly reduce reflectivity). I personally think its a hype, but there are some very high quality black media labels on the market, so they are not a bad buy.
A debate isn't my goal either. I would love to see an issue like this nailed down with solid proof to back it up. I haven't seen such thing online yet.
Ok, if the difference is strongly audioble, as you both describe the treble etc., then it should be visible in some sort of frequency analysis too? Now the only problem is to get the information loss-less from the CD player.
Can either of you use a digital output of the CD and record tracks back to PC for analysis?
|
|
Pio2001
Moderator
|
20. January 2003 @ 09:48 |
Link to this message
|
|
Rip your original and your copies into wav files, and use EAC to perform a compare wav between the two files. It must only show an offset difference (repeated/missing samples a the very end). Any "different samples" somewhere inside the file is data corruption and means that either the copy was not ripped properly, either it was not burned properly.
In the case of error free copies sounding different, so far, there is no explanation, it should be impossible, unless the player is bad and can't read the copy error free. It can be checked from the SPDIF output, if you have a soundcard capable of setting its clock in slave mode on the digital input in order to record the digital stream without performing ASRC between the player and the card clocks. Then you can compare the recording with a rip and see if there are differences (it won't tell if the errors are in the rip or in the copy, but it is easy to get an error free rip with EAC's "test and copy" method).
There is no evidence of the fact either. There has never been any objective report of this behaviour at all, neither from measuring instruments (distortion, frequency response...) nor from anyone's ears (positive blind ABX test, that would be an objective result).
|
|
Pio2001
Moderator
|
20. January 2003 @ 09:54 |
Link to this message
|
Correction : there is an explanation.
Bob Katz (http://www.digido.com , jitter) thinks that the pit/land jitter embedded into the copy burdens the servo engine that is spinnig the CD, that in turn burdens the power supply, that in turn unstabilize the onboard oscillator, that can't guarantee an optimal D-A conversion.
But it's not been measured, it is just an hypothesis.
Pio2001
|
|
Advertisement
|
|
|
|
mr_lemon
Junior Member
|
24. January 2003 @ 18:07 |
Link to this message
|
|
Hi,
I made a few TEST Cd's. I used Original music that was recorded by me, mixed to DAT, transfered to PC via S/PDIF and saved as WAV files.
I made two copies. First at 1x and the second at 8x.
I Ripped the files back to the hard drive and did the compare to the original WAV File.
Both 1x and 8x rips had the SAME errors, which are the missing samples mentions in Pio2001's message, and both are 0:00:00.0012 Longer then the original.
I also burned the wav as DATA, but the data disc file compare had NO errors at all. Data copys seem to be perfect, its when one does Audio copies that the above errors appear.
Im at a loss as to why it sounds different. I have read a lot on the Jitter argument, and it does sound like a good one. Usually pit/land jitter does not effect the bit compare, but does cause audible differences. The argument is that the CD player laser does need to work harder to read the pits/lands.
After all, CD's are really an analog media, with digital info on them. The laser is measuring the length of pits for its count of 1's (or 0's I forget which).
TOM...
|
